Abstract
OBJECTIVE: We report a gene × environment (health) study focusing on concurrent performance and longitudinal change in a latent-variable executive function (EF) phenotype. Specifically, we tested the independent and interactive effects of a recently identified insulin degrading enzyme genetic polymorphism (IDE rs6583817) and pulse pressure (PP; one prominent aging-related vascular health indicator) across up to 9 years of EF data in a sample of older adults from the Victoria Longitudinal Study. Both factors vary across a continuum of risk-elevating to risk-reducing and have been recently linked to normal and impaired cognitive aging. METHOD: We assembled a genotyped and typically aging group of older adults (n = 599, M age = 66 years at baseline), following them for up to 3 longitudinal waves (M interval = 4.4 years). We used confirmatory factor analyses, latent growth modeling, and path analyses to pursue 3 main research goals. RESULTS: First, the EF single factor model was confirmed comprising 4 executive function tasks and it demonstrated measurement invariance across the waves. Second, older adults with the major IDE G allele exhibited better EF outcomes than homozygotes for the minor A allele at the centering age of 75 years. Adults with higher PP performed more poorly on EF tasks at age 75 years and exhibited greater EF longitudinal decline. Third, gene × health interaction analyses showed that worsening vascular health (higher PP) differentially affected EF performance in older adults with the IDE G allele. CONCLUSION: Genetic interaction analyses can reveal differential and magnifying effects on cognitive phenotypes in aging. In the present case, pulse pressure is confirmed as a risk factor for concurrent and changing cognitive health in aging, but the effects operate differently across the risk and protective allelic distribution of this IDE gene.